Study on denoising of fractal signal based on Shannon entropy
Kai He, Shuxun Wang · 2003
In this paper, the wavelet coefficients' Shannon entropy of 1/f fractal signal and white noise at different scales is analyzed. Based on the proportion of wavelet coefficients' Shannon entropy of 1/f signal with that of signal polluted by noise at each scale, a new method of estimating 1/f fractal signal from added white noise is proposed in this paper. The simulation results show that this method is effective and practicable. Since it doesn't need to know the parameters of fractal signal and the statistical characteristic of added white noise in advance, this method is suitable in many situations.